Re: [gpsd-dev] OS X status?

[Eric S. Raymond]

Frank Nicholas <address@hidden>:
> On Jan 26, 2015, at 9:15 PM, Eric S. Raymond <address@hidden> wrote:
> > 
> > Greg Troxel <address@hidden>:
> >> regression tests are a complete disaaster on OS X 10.9.  They don't really
> >> look like 100% speed issues.   3.0 and 3.11 won't build.  Does anything
> >> work for anybody on mac?
> > 
> > Yikes.  I thought you were keeping an eye on that port.  I have no hardware
> > to test it here.
>
> I was a little active with gpsd in 2004-2005, with regards to OS X &
> Bluetooth based SIRF III receivers.  I haven’d touched it much since
> then, but still subscribe to and read the list.
>
> I have several Macs (OS 10.9 & 10.10) and am willing to do some
> testing to resolve these issues, or make my Macs available to
> someone via SSH.
>
> I’m no GIT expert, if someone can reply with a quick pointer to
> documentation on pulling the specific versions with issues, I’m
> willing to help troubleshoot.

Thank you very much for this generous offer.

The fastest way to get this done would probably be for me to ssh in
and blitz the problem.  OTOH, the best thing for the health of the
project would be for the Mac port to have an owner who does his own
light troubleshooting, invoking me only when there seems to be a core
code issue.

If you're willing to be that guy, here's what I recommend:

First, let's focus on 10.10.  Backporting to 10.9 shouldn't be
too difficult once that port is solid.

Second, the right version to pull is almost always current HEAD. An
unadorned "git pull" will do it.

The obvious first thing to do is just "scons check" and look at the
error messages.  What fails in your 10.10 environment?
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>